The poor health outcomes experienced by mothers and children in the developing world have been well known for decades, but practical, society-wide solutions remain elusive. By evaluating projects that seek to improve health through different channels in diverse contexts, this research program seeks to identify four such solutions. These projects are united not only by their goals, but in their rigorous methodology: controlled experiments of social programs. Project I examines the causal effect of fundamental drivers of demand for health services in Bangladesh. This project will trace the causal impact of education, age at marriage, and women's control over resources on women and children's outcomes by studying the long run implications of a program with three main parts: incentives to delay marriage, education support, and livelihood training. Project II assesses whether village-level incentive can foster collective action to improve demand for and access to health services. It examines the impact of a government program in Indonesia, Generasi, under which villages receive an annual block grant conditional on the village's performance on each of the 12 targeted health and education indicators. Project III examines how corruption can be controlled when subsidizing health product in a context of low demand. It will evaluate the impact of prices on the take up and usage of free insecticide treated bed nets and water purification kits and examine the impact of prices on corruption. It will then evaluate three interventions to reduce this corruption: auditing; market competition, and community empowerment. Project IV examines the impact of a catastrophic health insurance scheme offered by a microfinance organization to its clients in India, which covers hospital delivery. The project will study the impact of this program on the take up of institutional delivery, and on the impact of health shocks on household economic welfare. It will compare the impact of this program with that of a government program that gives women financial incentives for institutional delivery. PUBLIC HEALTH RELEVANCE: This research program seeks to identify cost-effective ways to improve maternal and child health through four rigorous evaluations. The first project measures the impact of delayed marriage for young women, the second evaluates a new approach to improve local health service delivery, the third identifies the best ways to deliver life-saving products, and the fourth measures the effect of insurance for institutional delivery.